update before release

.gitignore

@@ -1,175 +1,20 @@
-# Byte-compiled / optimized / DLL files
-__pycache__/
-*.py[cod]
-*$py.class
-
 # C extensions
 *.so
 
-# vim swap files
-*.swp
+# C++ files
+*.o
+*.d
 
 # Distribution / packaging
-.Python
 build/
-develop-eggs/
-dist/
-downloads/
-eggs/
-.eggs/
-lib/
-lib64/
-parts/
-sdist/
-var/
-wheels/
-share/python-wheels/
 *.egg-info/
-.installed.cfg
-*.egg
-MANIFEST
-
-# PyInstaller
-#  Usually these files are written by a python script from a template
-#  before PyInstaller builds the exe, so as to inject date/other infos into it.
-*.manifest
-*.spec
-
-# Installer logs
-pip-log.txt
-pip-delete-this-directory.txt
-
-# Unit test / coverage reports
-htmlcov/
-.tox/
-.nox/
-.coverage
-.coverage.*
-.cache
-nosetests.xml
-coverage.xml
-*.cover
-*.py,cover
-.hypothesis/
-.pytest_cache/
-cover/
-
-# Translations
-*.mo
-*.pot
-
-# Django stuff:
-*.log
-local_settings.py
-db.sqlite3
-db.sqlite3-journal
-
-# Flask stuff:
-instance/
-.webassets-cache
-
-# Scrapy stuff:
-.scrapy
-
-# Sphinx documentation
-docs/_build/
-
-# PyBuilder
-.pybuilder/
-target/
-
-# Jupyter Notebook
-.ipynb_checkpoints
-
-# IPython
-profile_default/
-ipython_config.py
-
-# pyenv
-#   For a library or package, you might want to ignore these files since the code is
-#   intended to run in multiple environments; otherwise, check them in:
-# .python-version
-
-# pipenv
-#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
-#   However, in case of collaboration, if having platform-specific dependencies or dependencies
-#   having no cross-platform support, pipenv may install dependencies that don't work, or not
-#   install all needed dependencies.
-#Pipfile.lock
-
-# poetry
-#   Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
-#   This is especially recommended for binary packages to ensure reproducibility, and is more
-#   commonly ignored for libraries.
-#   https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control
-#poetry.lock
-
-# PEP 582; used by e.g. github.com/David-OConnor/pyflow
-__pypackages__/
-
-# Celery stuff
-celerybeat-schedule
-celerybeat.pid
-
-# SageMath parsed files
-*.sage.py
-
-# Environments
-.env
-.venv
-env/
-venv/
-ENV/
-env.bak/
-venv.bak/
-
-# Spyder project settings
-.spyderproject
-.spyproject
-
-# Rope project settings
-.ropeproject
-
-# mkdocs documentation
-/site
-
-# mypy
-.mypy_cache/
-.dmypy.json
-dmypy.json
-
-# Pyre type checker
-.pyre/
-
-# pytype static type analyzer
-.pytype/
-
-# Cython debug symbols
-cython_debug/
-
-# PyCharm
-#  JetBrains specific template is maintained in a separate JetBrains.gitignore that can
-#  be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
-#  and can be added to the global gitignore or merged into this file.  For a more nuclear
-#  option (not recommended) you can uncomment the following to ignore the entire idea folder.
-#.idea/
 
 # ctags
 tags
 
-data
-examples
-umap
-nnd
+# Debug files
+Makefile
+trash.*
 callgrind.out.*
 gmon.out
-
-# cpp files
-*.o
-*.d
-
-# ignore this file
-.gitignore
-
-#
-trash.*
+nnd

CMakeLists.txt

@@ -13,26 +13,20 @@ set(CMAKE_CXX_FLAGS
 file(GLOB_RECURSE SRC_FILES src/*.cpp)
 
 # Library
-add_library(nndlib STATIC ${SRC_FILES})
+add_library(nndescent STATIC ${SRC_FILES})
 
 # Source code
-target_include_directories(nndlib PUBLIC ${CMAKE_CURRENT_SOURCE_DIR}/src)
+target_include_directories(nndescent PUBLIC ${CMAKE_CURRENT_SOURCE_DIR}/src)
 
 # Tests
-add_executable(test0 tests/test0.cpp)
-target_link_libraries(test0 PRIVATE nndlib)
-
 add_executable(test_distances tests/test_distances.cpp)
-target_link_libraries(test_distances PRIVATE nndlib)
+target_link_libraries(test_distances PRIVATE nndescent)
 
 add_executable(simple tests/simple.cpp)
-target_link_libraries(simple PRIVATE nndlib)
-
-add_executable(coil20 tests/coil20.cpp)
-target_link_libraries(coil20 PRIVATE nndlib)
+target_link_libraries(simple PRIVATE nndescent)
 
-add_executable(mnist tests/mnist.cpp)
-target_link_libraries(mnist PRIVATE nndlib)
+add_executable(faces tests/faces.cpp)
+target_link_libraries(faces PRIVATE nndescent)
 
 add_executable(fmnist tests/fmnist.cpp)
-target_link_libraries(fmnist PRIVATE nndlib)
+target_link_libraries(fmnist PRIVATE nndescent)

Makefile

@@ -1,6 +1,6 @@
 appname := nnd
 
-sources := src/dtypes.cpp src/rp_trees.cpp src/nnd.cpp tests/test0.cpp src/utils.cpp
+sources := src/dtypes.cpp src/rp_trees.cpp src/nnd.cpp tests/trash.test0.cpp src/utils.cpp
 objects := $(patsubst %.cpp,%.o,$(sources))
 depends := $(patsubst %.cpp,%.d,$(sources))
 

README.md

@@ -0,0 +1,66 @@
+# Nearest Neighbor Descent (nndescent)
+
+Nearest Neighbor Descent (nndescent) is a C++ implementation of the pynndescent library, originally written by Leland McInnes, which performs approximate nearest neighbor search. The goal of this algorithm is to construct a k-nearest neighbor graph quickly and accurately.
+
+## Background
+
+The theoretical background of NND is based on the following paper:
+- Dong, Wei, Charikar Moses, and Kai Li. "Efficient k-nearest neighbor graph construction for generic similarity measures." Proceedings of the 20th International Conference on World Wide Web. 2011.
+
+In addition, the algorithm utilizes random projection trees for initializing the nearest neighbor graph, based on the following paper:
+- DASGUPTA, Sanjoy; FREUND, Yoav. Random projection trees and low dimensional manifolds. In: Proceedings of the Fortieth Annual ACM Symposium on Theory of Computing. 2008.
+
+## Features
+
+- C++ implementation utilizing OpenMP for efficient computation
+- Support for dense matrices
+- Implementation of a subset of distance functions
+
+## Installation
+
+1. Clone the repository:
+
+```sh
+git clone https://github.com/brj0/nndescent.git
+cd nndescent
+```
+
+2. Build the project:
+
+```sh
+pip install .
+```
+
+3. Run the examples in `tests`. To build the dataset you should first run `make_test_data.py`
+
+## Performance
+
+On my computer, the training phase of nndescent is approximately 5-10% faster than pynndescent. Additionally, the search query phase is approximately 75% faster. Below is the output obtained from running tests/benchmark.py:
+
+### Benchmark test pynndescent vs nndescent
+Data set  | py train [ms] | c train [ms] | ratio | py vs c match | py test [ms] | c test [ms] | ratio | py accuracy | c accuracy
+----------|---------------|--------------|-------|---------------|--------------|-------------|-------|-------------|-----------
+faces     |         191.8 |        190.0 | 0.991 |         1.000 |       1631.6 |        20.5 | 0.013 |       1.000 |      0.999
+fmnist    |       13587.5 |      12935.1 | 0.952 |         0.997 |       6751.2 |      1757.2 | 0.260 |       0.978 |      0.978
+mnist     |       14187.2 |      12712.9 | 0.896 |         0.997 |       6664.2 |      1665.1 | 0.250 |       0.969 |      0.968
+
+The compilation time and the long numba loading time during import in Python for pynndescent are not taken into account here and are not required in nndescent.
+
+## Usage
+
+Please refer to the examples provided in the repository for instructions on how to use the NND library in your projects.
+
+## Contributing
+
+Contributions are welcome! If you have any bug reports, feature requests, or suggestions, please open an issue or submit a pull request.
+
+## License
+
+This project is licensed under the [BSD-2-Clause license](LICENSE).
+
+## Acknowledgements
+
+This implementation is based on the original pynndescent library by Leland McInnes. I would like to express my gratitude for his work.
+
+For more information, visit the [pynndescent GitHub repository](https://github.com/lmcinnes/pynndescent).
+